OBJECTIVES:We performed a meta-analysis of prognostic studies of patients with a Brugada ECG to assess predictors of events. BACKGROUND:The Brugada syndrome is an increasingly recognized cause of idiopathic ventricular fibrillation; however, there is wide variation in the prognosis of patients with the Brugada ECG. METHODS AND RESULTS:We retrieved 30 prospective studies of patients with the Brugada ECG, accumulating data on 1,545 patients. Summary estimates of the relative risk (RR) of events (sudden cardiac death [SCD], syncope, or internal defibrillator shock) for a variety of potential predictors were made using a random-effects model. The overall event rate at an average of 32 months follow-up was 10.0% (95% CI 8.5%, 11.5%). The RR of an event was increased (P < 0.001) among patients with a history of syncope or SCD (RR 3.24 [95% CI 2.13, 4.93]), men compared with women (RR 3.47 [95% CI 1.58, 7.63]), and patients with a spontaneous compared with sodium-channel blocker induced Type I Brugada ECG (RR 4.65 [95% CI 2.25, 9.58]). The RR of events was not significantly increased in patients with a family history of SCD (P = 0.97) or a mutation of the SCN5A gene (P = 0.18). The RR of events was also not significantly increased in patients inducible compared with noninducible by electrophysiologic study (EPS) (RR 1.88 [95% CI 0.62, 5.73], P = 0.27); however, there was significant heterogeneity of the studies included. CONCLUSIONS:Our findings suggest that a history of syncope or SCD, the presence of a spontaneous Type I Brugada ECG, and male gender predict a more malignant natural history. Our findings do not support the use of a family history of SCD, the presence of an SCN5A gene mutation, or EPS to guide the management of patients with a Brugada ECG.